A serological grouping scheme for Streptococcus sanguis has been developed. This scheme is based on gel diffusion analysis of Rantz and Randall (autoclave) extracts of S. sanguis and group H streptococci. The number of antigens identified in these extracts varied from one to five; one antigen (a) was found to be the group H antigen. Preliminary studies suggest that the group H antigen is a special type of teichoic acid. It gives a reaction of identity with the phenol water extracted teichoic acid but the latter does not show group specificity. It appears that the Rantz and Randall lipoteichoic acids (LTA) are either chemically distinct from the phenol water, LTA and thus represents a new class of polymers or that the Rantz and Randall procedures chemically alter the antigens. Strains which appear to have lost specific antigenic determinants (antigenic drift) relative to parent strain have also been observed. Both the specific nature of the loss and studies directed towards replacing the determinant by transformation reactions are underway. The taxonomic scheme based upon the immunochemistry of surface polymers has become valuable for tracing changes in the surfaces of these organisms. Such changes appear to play major roles in adherence of S. sanguis to tooth surfaces, salivary aggregation, plaque formation and antibiotic resistance. Studies are also concerned with determining which surface polymers are involved in cell to surface adhesion versus interbacterial aggregation. These studies may enable us to determine if modulation of serotypes in dental plaque is regulated by continuous reinfection or whether transformation reactions within plaque account for antigenic heterogeneity observed among plaque streptococci. The possibility that immunological reactions necessary to block plaque formation will depend upon "backbone" specificity rather than type specific antibodies is being explored.